Date: September 30, 2026
Time: 2:30pm ET
Chatbot, Workflow, or Agent?
Choosing the Right AI Architecture for Research
Build AI systems. More importantly, learn when not to.
AI agents are everywhere. But what actually makes something an AI agent? And when is an agent genuinely useful instead of unnecessarily complicated?
In this 3.5-hour hands-on virtual workshop, you’ll use practical research examples to explore how chatbots, workflows, and AI agents differ, then progressively build a simple research assistant in Google Colab.
Rather than focusing on a particular framework or vendor, you’ll learn the fundamental concepts behind modern AI systems and how to decide which approach is right for the problem you’re trying to solve.
What You’ll Learn
By the end of the workshop, you will be able to:
- Distinguish between chatbots, workflows, and AI agents
- Understand how tools extend AI capabilities
- Build a simple tool-using AI system
- Use document retrieval to ground AI responses
- Recognize where verification and human oversight are needed
- Identify research and administrative problems where agents add value
What You’ll Build
Using guided Google Colab notebooks, you’ll progressively build a simple research assistant.
Notebook 1: Chatbot vs. Workflow vs. Agent
Learn the key difference: who decides what happens next?
Notebook 2: Giving AI Systems Tools
Explore how AI systems select and use tools to complete tasks.
Notebook 3: Building a Research Assistant
Combine planning and document retrieval to produce a research briefing with verifiable citations.
Going Further: Memory and Human Oversight
An optional fourth notebook explores memory, task state, and human oversight for participants who want to continue learning after the workshop.
Who Should Attend?
This workshop is designed for faculty, researchers, research software engineers, research computing professionals, graduate students, and higher education staff interested in understanding how modern AI systems work and where they can be useful.
No previous experience with AI agents is required. Basic Python familiarity is helpful but not necessary.
No API Key Required
All core exercises run in Google Colab using a scripted mock mode, so no paid AI account or API key is required.
Participants will leave with reusable notebooks and example code, a practical framework for choosing between chatbots, workflows, and agents, and a certificate of completion from the MS-CC Agentic AI Virtual Workshop.
Speakers:
- Amanda Tan, Associate Director for Research Development, MS-CC
- Pavan Nara, AI Technical Specialist, MS-CC
Register Now
